UnicodeDecodeError even after setting `errors` to `ignore`

Question

There are UnicodeDecodeError even after I set the errors attribute of StreamWriter to ignore. Here is the code:

import codecs
import platform
if platform.system() == 'Linux':
    writer = codecs.getwriter('utf8')
    sys.stdout = writer(sys.stdout) # for writing to pipes
elif platform.system() == 'Windows':
    writer = codecs.getwriter('gbk')
    sys.stdout = writer(sys.stdout, 'ignore') # for writing to pipes

I run the script and get UnicodeDecodeError when print u'长城'.

Can anybody fix this for me ?

Update: the full errors is:

Traceback (most recent call last):
  File ".\main.py", line 42, in <module>
    print u'闀垮煄'
  File "D:\Program files\Python2.7.6\lib\codecs.py", line 351, in write
    data, consumed = self.encode(object, self.errors)
UnicodeDecodeError: 'ascii' codec can't decode byte 0xb3 in position 0: ordinal
not in range(128)

My OS is Windows 7.

What is `sys.stdout.encoding` set to before running the above? — Peter Gibson, May 13 '14 at 03:52
This isn't C++, you're not doing a cast. The expression `(writer)(sys.stdout)` is simply calling `writer` with `sys.stdout` as a parameter, and in my testing it appears to return whatever you gave it as the parameter. So you're essentially doing `sys.stdout = sys.stdout` which isn't going to help you. — Mark Ransom, May 13 '14 at 04:11
@PeterGibson 'cp936', I believe it means GBK, which is a Chinese encoding. — Shuai Zhang, May 13 '14 at 04:11
@MarkRansom But when I change it to `sys.stdout = writer(sys.stdout, 'ignore')`, it produces errors too. — Shuai Zhang, May 13 '14 at 04:18
@PeterGibson Suggested by Mark Ransom, I update the question. — Shuai Zhang, May 13 '14 at 04:22
What's the source code encoding odefined at the beginning of your script? (# -*- coding: gbk/utf-8/latin-1 -*-). And what's the real encoding of your script? — Leonardo.Z, May 13 '14 at 04:39
Good point @Leonardo.Z - try replacing any unicode characters in your script with their ASCII representations, eg `u'涓浗' => u'\u6d93\ue15e\u6d57'` — Peter Gibson, May 13 '14 at 04:46
@Leonardo.Z : I declare `#coding=utf8` in my python script, and the real encoding of the script is indeed utf8 — Shuai Zhang, May 13 '14 at 05:17
Please show your real code. I am sure line 39 in your script is not `print u'长城'` — Leonardo.Z, May 13 '14 at 05:23
@Leonardo.Z Really sorry, it was `print u'中国'` in my local script. I update the error message. Sorry... — Shuai Zhang, May 13 '14 at 05:26

UnicodeDecodeError even after setting `errors` to `ignore`

0 Answers0